Author: Ferdinand S. Mosca
Date: 03:22:50 06/08/01
Go up one level in this thread
On June 08, 2001 at 05:57:27, Mike S. wrote: >On June 08, 2001 at 01:53:58, Ferdinand S. Mosca wrote: > >>Hello Bob, >>(...) >>[D]8/8/1n6/2b2B2/5B1k/3p4/5P2/5K2 b - - 0 62 >>Here Crafty played 62...Bxf2?? with an evaluation of 0.0 >>on P3 668mhz. > >I have tried Crafty 18.08 with 4-piece tbs; it plays this move too (some other >programs don't). > >62...Bxf2! is a Master's move: > >1. The pawn d3 is lost, black cannot go for a win I agree and we will go for a draw here. >2. White might try to win with his pawn what if at certain point the knight will take the pawn and not the bishop? And goes for BB vs. B which draws. >3. BB vs. N should need *more than 50 moves* - obviously Crafty knows that - so >Bxf2 "simply" draws! Interesting, but this maybe unclear as Comet will first hunt the knight which is what actually happened in the game!! > >A good example, how endgame knowledge in a chess program can lead to surprising >"master-like" moves (which are not boring and less clear, i.e. like ...Nd5 (?)). Nd5!?, at least there is resistance, Comet's position is not very threatening yet, its too early to give up the Bishop. > >The only question which may arise IMO is, if Crafty can be self-confident to >keep the N-BB draw without the 5-piece tbs, or without tbs at all, for the >necessary 50 moves (it plays Bxf2 without tbs too). > >Regards, >M.Scheidl
This page took 0 seconds to execute
Last modified: Thu, 15 Apr 21 08:11:13 -0700
Current Computer Chess Club Forums at Talkchess. This site by Sean Mintz.